基于DEAFCR算法的非线性系统模型参数辨识

Parameter identification of nonlinear system model based on DEAFCR algorithm

针对非线性模型参数辨识困难和不准确的问题,提出一种基于改进的差分进化算法的辨识算法。通过建立寿命机制,根据寿命值,动态调整缩放因子和交叉率,在算法初期保持多样性来避免早熟收敛,在后期保留优质解,加快收敛速度。为验证改进算法的性能和实用性,用典型测试函数进行对比测试,并辨识非线性传递函数模型和Hammerstein模型,试验结果表明改进的算法收敛速度快,辨识精度高,对非线性系统参数辨识有效可行。

Aiming at the difficulty and inaccuracy of nonlinear model parameter identification, a new identification algorithm based on improved differential evolution algorithm is proposed.The life mechanism is established.According to the life value, the differential weight and crossover rate are dynamically adjusted.At the beginning of the algorithm, diversity is maintained to avoid premature convergence,and the high quality solution is retained at the later stage to accelerate the convergence speed. In order to verify the performance and practicability of the improved algorithm,several typical test functions were compared and tested,and the nonlinear transfer function model and Hammerstein model were identified.The experimental results show that the improved algorithm has high convergence speed,high identification accuracy, and nonlinear system parameters.The identification is effective and feasible.